Use of antiretroviral (ARV) therapy to suppress HIV and allow immune reconstitution has shifted mortality from opportunistic infections to complications of chronic infections and ARV therapy. A significant and potentially fatal adverse effect is hepatotoxicity. Recent rapid expansion of ARV therapy in developing countries has the potential to reverse the AIDS mortality currently devastating these areas but drug toxicity may present a barrier to program success. One known risk factor for hepatotoxicity, hepatitis B (HBV), is hyper-endemic in parts of Africa. In addition to HBV, multiple other known and potential risks for hepatotoxicity may exist including alcohol, traditional medicines, and tropical diseases. We propose to test the hypothesis that the incidence of ARV-related hepatotoxicity is greater in Africa than in North America and that risk factors differ. This study will measure incidence of hepatotoxicity in an African ARV therapy cohort and identify risks for hepatotoxicity by matching cases with controls and testing banked serum and a administering a structured interview. It will be conducted within a large and on-going ARV cohort established as a collaborative project between a South African industrial conglomerate and Johns Hopkins University. [unreadable] [unreadable]